In the Washington Post last week, housing columnist Kenneth Harney took an interesting look at mortgage loans guaranteed by the Department of Veterans Affairs.
VA lending presents a puzzle. How can a program where 91% of borrowers make no down payment, where borrowers have lower average credit scores than at Fannie Mae and Freddie Mac, and where borrowers' total debt load in some cases exceeds half of their monthly incomes, have a relatively low default rate?
According to the article, the latest 90-day default rate for VA lending was 2.2%, less than half the rate at the Federal Housing Administration.
So what's the VA's secret? Here's what Harney wrote:
Michael Frueh, the VA program's acting director, says the key to the agency's quiet success is its almost paternalistic emphasis on servicing its 1.5 million borrowers — moving early and quickly to intervene at the slightest hint of payment problems.
"At the end of the day, we are veterans' advocates," he said in an interview. "We exist solely to help them," not only to afford to finance their homes but to remain in them.
This explanation makes a lot of sense to me — and it also ties together with a somewhat obscure historical precedent. During the Great Depression, which was the last time that housing prices took a sharp downturn across the entire country, mortgages serviced by the federal government performed a lot better than one might have expected.
Prior to joining American Banker, I worked on the staff of the Congressional Oversight Panel for TARP. For a report on the Obama administration's response to the foreclosure crisis, I researched a Depression-era program called the Home Owners Loan Corporation, or HOLC. Our goal was to determine whether the federal government's response to the foreclosure crisis of the 1930s held any lessons for policymakers today.
(You can read the Congressional Oversight Panel's conclusions about HOLC here. See pages 90-92 and 111-122 of the PDF.)
Here's how HOLC worked. The government bought defaulted mortgages from private lenders and refinanced them in order to make the monthly payments more affordable. This turned the U.S. government into an enormous mortgage lender and servicer.
By the time HOLC shut down in the early 1950s, it had a pretty impressive record. Despite some large upfront costs, the program lost far less money than was expected. And over time default rates on HOLC loans fell rather dramatically. Obviously, improvements in the U.S. economy were one reason why. But another potential reason, we concluded, had to do with HOLC's loan-servicing practices.
… the HOLC went to great lengths to keep homeowners in their houses. At its peak, the HOLC had about 20,000 employees and offices in 48 states. In servicing loans, the HOLC relied heavily on personal contact aimed at helping distressed homeowners. Servicing practices varied by state and over time, but there were some common themes. In the second and third months of delinquency, the HOLC would insert special notices with the homeowner's monthly bill. If the homeowner was unresponsive, a form letter followed. Next came a personal letter. If there was no response, a HOLC staffer would make an in-person visit to the home. In some cases, HOLC employees helped homeowners or their relatives find jobs, and to collect insurance claims, unpaid debts, and pensions. HOLC employees even suggested ways of finding tenants or foster children to defray the homeowner's monthly mortgage payment. C. Lowell Harriss, the author of a 1951 book about the HOLC, wrote: "The assistance given by the HOLC's service representatives is difficult to summarize adequately. The closest parallel, perhaps, is found in the social worker's helping individuals and families adjust to their own problems and to the community around them." All of this stands in marked contrast, of course, to the highly automated, impersonal mortgage servicing practices of today.
Which brings me to HAMP. A lot of ink has been spilled about the reasons for the program's failures. In my view, the biggest problem is that it's being carried out by loan servicers whose main goal is not keeping families in their homes. By offering a few thousand bucks to servicers that agree to modify mortgages, the program attempts to change the incentives for servicers. But the program's dismal performance suggests that the financial incentives being offered aren't big enough.
As the Congressional Oversight Panel concluded:
Whereas the HOLC was a purely governmental effort, HAMP, by its design, relies on private servicers to carry out public-policy aims. This creates a principal-agent problem — there may be a divergence of interests between Treasury and the servicers it is paying — which could be problematic even if HAMP were a mandatory program. Unfortunately, the voluntary nature of HAMP exacerbates the problem, because Treasury has few tools available to ensure that servicers fulfill their obligations under the program.
The U.S. government's goal when it established HOLC was to keep as many people in their homes as possible. That's also what the VA is trying to do. It's not a mission that can easily be outsourced.